The influenza virus is one of the most genetically mutable pathogens of humans. Our immune response places substantial pressure on the virus, resulting in rapid turnover of genetic lineages from year to year. Our immune response also contains a record of the strains we have been infected with in the form of strain-specific antibodies. Though some prospective studies of influenza viruses have been conducted, little has been done to investigate the distribution of immunological profiles in space or time. These patterns present a rich data source for investigation of the impact of human social interactions and demographics on influenza transmission, key determinants of the ecology of the influenza virus. The urban and rural areas in and near Guangzhou and Dongguan in Guangdong province, China, provide a unique opportunity. The project team will conduct a study of the immune landscape of influenza, linking that immune landscape, at the individual-level, to data on household structure, travel behavior and social networks. As well as being a key human population for recent emergent infections (SARS and H5N1) the study population also exhibits important heterogeneities in social structure which will allow us to investigate novel and interesting hypotheses. For example, both population density and average socio-economic status vary massively over relatively small distances. The project team will use previously-validated questionnaires on household structure, travel and social contacts to gather data from each member of 20 households in 50 small communities on two separate occasions (one year apart). They will also gather blood samples on both occasions. Viral neutralization assays will be conducted for all blood samples against a representative set of 12 strains from the 10 years prior to the start of the project. These social structure and serological data will be used alongside mathematical models of infectious disease transmission to help answer fundamental questions about how human behavior influences disease transmission. For example, does influenza skip some small towns on a given year but infect all geographical areas of large cities, do children bring much more infection into households than adults, and can we predict that some geographical areas might be harder hit in future seasons using serological data from past seasons? The mathematical models used to help answer these questions will vary from simple differential equation models to large-scale individual-based simulation models. As well as being a valuable research tool, large simulation models may prove to be valuable public health tools when they can be accurately calibrated using serology and social structure data of the kind the team will gather in this project. PUBLIC HEALTH RELEVANCE: Although the work proposed here will have broader impact beyond the publication of high quality peer-reviewed scientific articles, it should be remembered that the published results themselves will be of direct use to public health policy makers in the event of a pandemic. This study will provide more accurate estimates of the likely epidemiology of an emergent pandemic of influenza in the human population of southern China than any previous studies (in any population). Key stakeholders such as the Center for Disease Control in Guangzhou and the Center for Health Protection in Hong Kong will be made aware of these results during meetings held specifically for them.